Commentary on: Peterson JL, Hickman MJ, Strom KJ, Johnson DJ. Effect of forensic evidence on criminal justice case processing. J Forensic Sci 2013 Jan;58 Suppl 1:S78-90

摘要

What should the unit for the analysis be? Here are the two main options: (i) one could take all crimes (i.e., aggregate the crimes) as the unit. This approach makes use of patterns of covariation in criminal case processing and its supposed causes to provide evidence about causal influences or (ii) one could analyze the case-processing outcomes of each crime separately with an eye toward identifying differences among them in the predictors of case-processing outcomes.Peterson et al.'s choice of analysis was to aggregate all crimes into a unit unto itself. This choice is highly problematic because the nature, context, and elements of each crime vary greatly from one to another: The type and quantity of forensic evidence collected at crime scenes; the likelihood of submission and ultimately examination of forensic evidence; the use of weapons, particularly guns; injuries -to victims; and the offender-victim relationship all vary significantly across crimes. An analysis that treats crime as a unitary entity will fail to see these nuances. An important question that an aggregate analysis cannot address is: Were the predictors of case outcomes, specifically the influence of forensic evidence, fairly uniform from one crime to another? Or were there sharply divergent patterns across crimes? Simply including an ordinal variable of crime seriousness (the Peterson et al.'s solution) does not provide answers to this question.